An experimental study of fault growth in a 2D Biaxial apparatus

Placing constraints on the geometry and growth of faults has significant implications for the management of resources in the subsurface; faults are widespread structures and can either compartmentalise subsurface reservoirs or provide favourable fluid migration pathways.

There are two widely accepted models of fault growth: the isolated fault growth model and the constant length fault growth model. These models, largely derived from field, seismic, and analogue modelling data, both describe three stages in fault growth: (1) propagation of fault segments, (2) slip accumulation on fault segments, and (3) segment interaction and linkage. However, they differ in how the initial fault segments interact, whether they are kinematically dependent, and how rapidly their full length is established. The debate is currently still open on which model best describes natural faults, and what geological controls favor one model over the other.

To further address the topic of fault growth, we investigate the different stages of growth through two sets of experiments. First, loading experiments are performed on intact samples of a rock analogue material to track both the propagation of the fracture and the displacement accumulation to test which of the two fault growth models most accurately describes the initial stages of fault propagation. Second, loading experiments are performed on samples with pre-cuts to replicate realistic fault segment geometries, to track fault tip migration and displacement partitioning during the linkage stage of fault growth, to test geometrical controls on the process of fault linkage.

The samples are made of a rock analogue material capable of accommodating displacement gradients through ductile processes, similar to those observed in natural rocks over geological timescales. This material is cohesive and allows the creation of pre-cuts to replicate fault segment geometries. Loading experiments are conducted in a biaxial apparatus at low strain rates, coupled with an interferometric technique using an optical bench to obtain speckle patterns. These speckles are employed to track in high resolution the increase in the length of the fault using Diffusive Wave Spectroscopy (DWS). In addition, we concurrently use the same speckle patterns to track displacement along the fault using Digital Image Correlation.